* boost::thread mostly gone... still need to get rid of once
everything compiles
* replace boost::call_once
* remove link-time dependency on boost::thread
* first pass at using async
* switch to using async everywhere
* Make sure EmbedMultipleConfs is deterministic for very large seeds and a seed of 0.
Fixing two distinct bugs:
- A randomSeed of 0 didn't use '0' as the seed even though the
documentation said that it should.
- Very large seeds could overflow the 'int' representation during
calculation of a new internal seed. This would make
EmbedMultipleConfs non-deterministic even though the user provided a
valid seed.
* Adding a C++ test for case pull request #1635.
* version 2 of ETKDG
* setting default and changing to unsigned int
* ETversion 2 only via EmbedParams available
* Some refactoring so that the DG functions that take the parameter object are the primaries.
This should allow new parameters to be added without having to modify the API
* test and wrapper changes to go along with that
* update cookbook to use etkdg
* add a test (currently fails, of course)
* backup, not really working
* Fixes#1240
* a bit more parameter tweaking to get some more structures to embed
* initial version; no actual testing yet
* start on tests
* rename seed->randomSeed; add tests
* add support in java wrapper
* add support in python wrapper
* initial version; no actual testing yet
* start on tests
* rename seed->randomSeed; add tests
* add support in java wrapper
* add support in python wrapper
* yapf
* document the parameter object;
introduce pre-defined const objects for ETDG, ETKDG, and KDG
* this is why we write tests
* swig did not like the initialization in the header file
* passes all tests, but is still not 100 percent there
* generalize _centerInVolume for possible future use
* better testing of tetrahedral centers
update tests
* only test ring atoms
* handle 4-coordinate n too
* add a volume check; not all tests pass
* turn off debug printing
* rearrange the order of tests.
if this is not done, we get a seg fault when the github55 test runs.
the whole thing needs to be run under valgrind to track this down
* clear up a memory leak
* a bit more documentation
add a constant for one of the threshold values
* get more permissive on the energy tests.
Only do the extra tetrahedral tests for atoms in at least two rings.
* add a DEBUG_EMBEDDING option to make tracking down failures easier
* enable better debugging when the flag is set
* remove a FIX
* remove some debug printing from the tests
* enforce planarity
* increase the force constant for the impropers
* force constant adapted
* reduced tolerance for planarity and force constants changed for some torsions
* tolerance for planarity increased a bit again
* cerr outputs removed
* planarity tolerance increased
* boost log added in planarity check
o rdkit gains a RDKit::common_properties namespace that contains common string value properties
o Dict.h and below gain getPropIfPresent that attempts to retrieve a property and returns
true/false on success or failure. This is used to optimize access.
o rdkit learns how to pass property keys by reference, not value.
A new namespace has been added to RDKit, common_properties
that contains the std::string values for commonly used
properties. This helps to avoid typos in string values
but also avoids a creation of std::strings from character
values. All accessors (has/get/clear and getPropIfPresent) now pass
the key by reference.
Additionally, getPropIfPresent removes the double lookup
of hasProp/getProp which can be a significant speedup
in the smiles and smarts parsers (10-20%)
remove const specifiers from return values in DataStructs
add initial pass at query-query matching for Atoms and Bonds
fix and test for sf.net issu 2738320
add Mol.AddRecursiveQuery() method to Mol
instead of using the property map interface.
A nice side-effect is that the wart of having to use property maps to loop over bonds or atom neighbors
is now gone.
This potentially breaks lots of client C++ code.
